[BitTorrent] Back to Merkle Hash Trees...

Olaf van der Spek OvdSpek at LIACS.NL
Fri Feb 4 18:47:53 EST 2005


> On Fri, 04 Feb 2005 00:38:59 -0000, sh4dowmatter <mgp at ucla.edu> wrote:
>
>> This must all seem utterly trivial, so I just want wondering if the
>> powers that be, whether it was originally Merkle or someone else, said
>> that "all Merkle trees shall be formed breadth-first, and not as a
>> full binary tree." If I try pursue this idea above, can I call it a
>> Merkle hash tree?
>
> Either way, it is a merkle hash tree. The question is if it complies
> to that THEX format. I'd still say so, because in point 3.3 they say
> "Normal breadth-first serialization is the recommended manner in
> which to serialize the hash tree.". Note the *recommended*.
>
> The structure of the Hash Tree depends on the order in which you
> combine the leaves. There's a lot of variations possible. If your
> program wants to build hash trees your way thats no problem. But
> remember that it should be able to read *any* hash tree correctly.

Isn't there just a single way to combine two hashes? First the left, then 
the right?

> HTH,
> Kosta
> who still doesnt understand where the big advantage over a list
> of hashed blocks is, anyways

Tiny .torrent size, smaller verification granularity, compatibility with 
other merkle tree hashes. 



 
Yahoo! Groups Links

<*> To visit your group on the web, go to:
    http://groups.yahoo.com/group/BitTorrent/

<*> To unsubscribe from this group, send an email to:
    BitTorrent-unsubscribe at yahoogroups.com

<*> Your use of Yahoo! Groups is subject to:
    http://docs.yahoo.com/info/terms/
 





More information about the BitTorrent mailing list